IRC log of #maemo-ssu for Saturday, 2012-10-27

*** nox- has joined #maemo-ssu00:00
*** arcean has quit IRC00:12
*** arcean has joined #maemo-ssu00:14
*** arcean has quit IRC00:25
*** arcean has joined #maemo-ssu00:32
*** M13 has joined #maemo-ssu00:39
*** Adeon is now known as Adeon\SIGSEGV00:43
*** arcean has quit IRC00:45
*** arcean has joined #maemo-ssu00:46
*** toxaris has quit IRC00:55
*** arcean has quit IRC01:26
*** andre__ has quit IRC01:29
*** M4rtinK has quit IRC01:32
*** dafox has quit IRC01:54
*** MrPingu has quit IRC01:54
*** dafox has joined #maemo-ssu01:57
*** andre__ has joined #maemo-ssu02:17
*** andre__ has joined #maemo-ssu02:17
*** dhbiker has joined #maemo-ssu02:22
*** dafox has quit IRC02:30
*** dafox has joined #maemo-ssu02:31
*** BCMM has joined #maemo-ssu02:40
*** dafox has quit IRC02:52
*** dafox has joined #maemo-ssu02:53
*** dafox has quit IRC03:00
*** BCMM has quit IRC03:15
*** dhbiker has quit IRC03:24
*** arcean has joined #maemo-ssu03:26
*** arcean has quit IRC03:55
*** amiconn has quit IRC05:12
*** amiconn_ has joined #maemo-ssu05:12
*** amiconn_ is now known as amiconn05:12
*** DocScrutinizer05 has quit IRC06:03
*** DocScrutinizer05 has joined #maemo-ssu06:03
*** nox- has quit IRC06:11
*** Jaded has quit IRC06:16
*** Jade has joined #maemo-ssu06:21
*** Jade has joined #maemo-ssu06:21
*** SpacedOut has left #maemo-ssu06:29
*** amiconn has quit IRC06:53
*** amiconn_ has joined #maemo-ssu06:53
*** amiconn_ is now known as amiconn06:53
*** M13 has quit IRC08:26
*** M13 has joined #maemo-ssu09:22
*** dhbiker has joined #maemo-ssu09:34
*** andre__ has quit IRC11:08
*** toxaris has joined #maemo-ssu11:28
*** toxaris has quit IRC11:43
*** toxaris has joined #maemo-ssu11:59
*** M4rtinK has joined #maemo-ssu11:59
*** dhbiker has quit IRC12:28
*** NIN101 has joined #maemo-ssu12:54
*** freemangordon has joined #maemo-ssu12:56
*** dhbiker has joined #maemo-ssu13:06
*** iDont has joined #maemo-ssu13:20
*** ekze has joined #maemo-ssu13:21
*** arcean has joined #maemo-ssu13:37
*** arcean has quit IRC13:39
*** arcean has joined #maemo-ssu13:40
*** arcean has quit IRC14:20
*** arcean has joined #maemo-ssu14:24
*** dafox has joined #maemo-ssu14:49
*** _rd has joined #maemo-ssu15:04
*** BCMM has joined #maemo-ssu15:12
*** _rd has quit IRC15:16
*** LaoLang_cool has joined #maemo-ssu15:23
*** arcean has quit IRC15:45
*** iDont has quit IRC15:46
*** arcean has joined #maemo-ssu15:46
*** crashanddie has joined #maemo-ssu15:53
*** crashanddie has joined #maemo-ssu15:53
*** crashanddie has quit IRC16:05
*** Mihanizat0r has joined #maemo-ssu16:14
*** M13 has quit IRC16:16
*** freemangordon has quit IRC16:25
*** freemangordon has joined #maemo-ssu16:25
kerioyo freemangordon16:39
*** Mihanizat0r has quit IRC16:53
*** LaoLang_cool has quit IRC17:20
*** dafox has quit IRC17:25
*** dafox has joined #maemo-ssu17:26
*** jonwil has joined #maemo-ssu17:28
*** _rd has joined #maemo-ssu17:31
freemangordonkerio: hi17:45
keriosup17:46
freemangordonoil prices?17:46
kerioooh, neat17:49
kerioanyway, moar thumb plskthx17:50
freemangordonwill try to put some more packages in the repo soon17:53
kerioyay17:53
keriopulse/tracker?17:53
keriothat would be cool17:53
freemangordonwithout -mp thing initially, I wan't them tested by a couple of brave souls/guinea pigs before doing -mp thingie17:53
keriofreemangordon: what's your opinion of sqlite3 from marmistrz?17:53
freemangordonkerio: do I have one?17:54
freemangordon(opinion)17:54
kerioi mean, it's something that's not in cssu17:54
keriowould you consider compiling it?17:54
kerioand shipping it, i s'pose17:54
freemangordonkerio: I already ansered that quiestio17:54
freemangordon*answered17:54
kerio...although technically for "system" packages it's not the same thing, you can put it on the repo without adding it to the metapackage17:55
freemangordonwon't put newer sqlite 3 until it is not in 'main' CSSU17:55
kerioshame17:55
kerio:c17:55
freemangordonwhy?17:55
freemangordonyou have some problems with stock sqlite3 version?17:56
kerioi'm using the more recent one, and it feels a bit faster17:56
keriotracker uses sqlite17:56
freemangordongcc 4.6.2?17:56
freemangordonor 4.6.1?17:56
*** arcean has quit IRC17:56
keriofreemangordon: how do i check?17:57
freemangordonNFC17:57
freemangordonbut IIRC marmistrz's has compiled it with gcc4.6.x17:57
keriohm17:58
keriothen it's ok, i suppose17:58
freemangordonkerio: BTW - "3.6.14-1maemo4+0m5+0cssu0+thumb0"17:58
freemangordonthis is in -thumb17:58
freemangordoncan you check your?17:58
freemangordon*yours17:58
keriowhich package?17:59
keriosqlite3, libsqlite3-0?18:00
freemangordonboth18:00
keriooh, the version is the same18:00
kerio3.7.7-2+maemo2+0m6maemo018:00
kerioi wonder if the thumb version would be faster18:00
freemangordonshould be18:00
freemangordondifference between 47-linaro and 4.6-upstream for arm is noticeable18:01
keriohm18:01
freemangordonok, going to thumb-compile pulseaudio18:02
*** dafox has quit IRC18:05
*** arcean has joined #maemo-ssu18:12
*** jonwil has quit IRC18:14
DocScrutinizer05thumb-compile pulseaudio | tee -a /etc/bullshit >/dev/null18:25
DocScrutinizer05or terse:18:25
DocScrutinizer05~nuke PA18:25
* infobot prepares 100 missle silos, and targets them at PA ... B☢☢M!18:25
keriooh, i thought you were going to say thumb-compile pulseaudio >> /etc/bullshit18:25
DocScrutinizer05alas we're obviously stuck to that poettering PA BS, since it's way too entangled with whole maemo system18:26
DocScrutinizer05so getting it thombified could only improve shit18:27
keriowhat's the alternative?18:27
kerioalsa+dmix?18:27
freemangordonexactly, we'll have better PA shit :D18:27
DocScrutinizer05yeah18:28
DocScrutinizer05better shit >> shit18:28
DocScrutinizer05no sarcasm, modulo at PA in general18:28
DocScrutinizer05kerio: see last post, 3 lines up: there's no alternative18:28
keriono, i meant when hypothetically redesigning everything18:29
DocScrutinizer05there'd be the alternative of pure ALSA with a few small extensions I designed, if we'd stop a new OS out of the ground18:29
DocScrutinizer05stomp*18:29
keriowell, let's do that18:31
kerio>:D18:31
DocScrutinizer05and it would work way more seamlessly with any generic ALSA audio enabled app, without special freakin tweaks to deal with shite like alsaped etc18:31
DocScrutinizer05all it needs is a .alsarc device definition for each app using audio, to define an ALSA device exclusively for that app18:32
DocScrutinizer05so the whole friggin properties like audio policies etc would get defined implicitly by the device definition18:33
keriohm, is that efficient enough?18:33
DocScrutinizer05each app had their own volume control18:33
DocScrutinizer05o.O18:33
DocScrutinizer05efficient?18:33
DocScrutinizer05in which way?18:33
DocScrutinizer05is PA efficient??18:33
kerioa new ALSA device for each program18:33
DocScrutinizer05errr18:34
keriounless i misunderstood18:34
DocScrutinizer05you're aware what's a virtual ALSA device? it's nothing more than just a definition of the stack of lib plugins and their paramenters to use by the particular app, executed in app18:35
DocScrutinizer05's addr space18:35
DocScrutinizer05it's absolutely no difference whether every app is using "ALSA:default" or each app has their own "ALSA:myaudiodev-$appname"18:36
keriook, so it's efficient18:36
keriocool18:36
DocScrutinizer05performance-wise18:36
DocScrutinizer05you might want to check out http://tech.groups.yahoo.com/group/twinklephone/message/1731 and http://people.openmoko.org/joerg/ALSA/ACI/18:41
DocScrutinizer05particularly POC in http://people.openmoko.org/joerg/ALSA/ACI/acihooklib/18:42
DocScrutinizer05which I invented18:42
DocScrutinizer05it's a friggin useful ALSA plugin to basically execute arbitrary commands whenever an arbitrary app opens (or closes) the dedicated ALSA device18:43
jon_yspeaking of VOIP, has anybody tried to pipe it trough the voice modem? :)18:43
jon_yusing my voice modem as a pots modem :)18:43
DocScrutinizer05you could switch on your home stereo PA when sip-phone ringer opens riger device, so you hear phone ringing even while you're in the garden18:44
DocScrutinizer05of course you also can setup arbitrary mixer setup to accomodate the "policies" of ringer or whatever the device18:44
kerio"I HAVE THE SOFTWARE 21.2011.38.1 ON MY N900.when i install from CSSU the software open a window of Xterm and he do the operation . after that he said me that is an error i don't remember where and he talk me to press a key to get out . ok after that i install the update . after that the orientation LANDSCAPE - Horizondat - vertical doesnt work . same thing in the applications ... i dont know what to do. i will REINSTALL ALL THE MAEMO and i will not in18:44
keriostall ever CSSU .i cant take a software if is not working.better get him out.thanks"18:44
keriowhat the fuck18:44
keriothis shit right here is the reason i don't follow TMO too closely18:45
DocScrutinizer05lart this moron, fire, forget18:45
jon_yihavethesoftware21.2011.38.1onmyn900wheniinstallfromcssu....18:45
jon_ysounded like that to me :)18:46
DocScrutinizer05URL?18:46
DocScrutinizer05I feel like answering ;-P18:46
freemangordonDocScrutinizer05: I like the idea, gimme a minute18:47
keriohttp://talk.maemo.org/showthread.php?p=128479818:47
jon_yheh, ihavethesoftware21.2011.38.1onmyn900wheniinstallfromcssu...sobsobsob18:48
*** Pali has joined #maemo-ssu19:00
BCMM"an error i don't remember where" - "well there's a fix but i don't remember what it is"19:01
*** toxaris has quit IRC19:08
*** andre__ has joined #maemo-ssu19:23
*** andre__ has joined #maemo-ssu19:23
freemangordonkerio: PA is in thumb repo, upgrade if you dare :P19:29
freemangordon160k more free in rootfs19:29
ShadowJKfreemangordon; been doing anything more on ramzswap?19:31
freemangordonI suspended for a while, until it is clear if it makes sense to continue19:31
keriofreemangordon: do i dare? D:19:32
keriomeh, i'm taking a full backup first19:32
freemangordonhmm, telefony still works19:32
freemangordon*telephony19:33
freemangordonaudio playback too, so it should be ok19:33
kerioit's just a recompile, isn't it19:33
ShadowJKfreemangordon; well, would be interesting to see how much of swapin comes from the compressed part with your current version19:34
freemangordonkerio: yes19:36
freemangordonplus luf's patches for BT19:36
freemangordonShadowJK: yeah, I was thinking to remove compression.19:37
freemangordonthat way we'll have flash-optimized swap19:37
keriofreemangordon: a cheap compression is probably better anyway19:37
keriounless the NAND is *really* fast19:37
freemangordonbut want first to issue a new update in -thumb as it passed a month or so since the last one19:37
freemangordonkerio: no, it is not19:38
Palikerio, I'm here19:38
freemangordonn900 has not enough RAM for that to be effective19:38
keriofreemangordon: :(19:38
Palikerio, what did you asked yesterday?19:38
keriofreemangordon: should i reboot?19:45
keriomeh, i'll reboot19:45
freemangordonkerio: yes19:45
freemangordonkerio: make sure you've upgraded 7 packages19:46
kerioi just did apt-get upgrade bro19:46
kerio:P19:46
keriofreemangordon: works19:51
DocScrutinizer05Pali: you asked me to let luf and freemangordon know sth. Now you can do that your own19:58
freemangordonkerio: yeah. lets see how will that affect battery life19:58
PaliDocScrutinizer05, luf already received messages :-)19:59
DocScrutinizer05freemangordon: i'd be most interested in CPU hogging of PA before and after thombification19:59
DocScrutinizer05freemangordon: battery life is a secondary function of that20:00
freemangordonDocScrutinizer05: :nod:20:00
freemangordonthough aiui the one that uses CPU is in the closed part20:00
keriofreemangordon: when on speaker, though20:01
DocScrutinizer05freemangordon: yeah, prolly XPROT20:01
keriocan't we get some good speakers to install?20:02
freemangordonkerio: ?20:02
DocScrutinizer05though, iirc PA mixing is a hogging greedy bastard as well20:02
DocScrutinizer05kerio: hardly20:02
kerioat least speakers that won't break if you look at them wrong20:02
freemangordonDocScrutinizer05: any idea how to measure that?20:02
DocScrutinizer05nifty20:03
DocScrutinizer05kerio: speakers are fine for their size and compartment20:03
kerioDocScrutinizer05: they break!20:03
DocScrutinizer05the problem is missing low-cut filter in whole audio chain, and speakers of that size generally can't deal with frequencies below ~150Hz20:04
DocScrutinizer05that's basically what XPROT is suspected to accomplish: cut out low frequencies20:04
DocScrutinizer05it's basically a hw bug20:04
kerioyeah, let's get some new hw without that bug then :)20:05
DocScrutinizer05the 180° opposite of what we had on FreeRunner ;-P20:05
DocScrutinizer05there we had a lowcut in hw that didn't allow freq < ~1kHz on headset :-o20:06
freemangordonDocScrutinizer05: we can lock the frequency to lets say 250 and check PA CPU usage20:06
freemangordonplaying one and the same song20:06
DocScrutinizer05sure20:06
freemangordonan both devices20:06
freemangordon*on20:06
freemangordoni'll do the test tomorrow20:06
DocScrutinizer05cool :-)20:06
kerio"we're gonna play to one song, and one song only"20:07
Skrybtw what resampler PA is using in fremantle?20:08
DocScrutinizer05hm?20:08
DocScrutinizer05isn't the resampler aka rate-conversion an immanent part of PA as well as ALSA?20:09
DocScrutinizer05anyway PA 'mixer' been notorious to do "(0 + 0) / 2" 44k/s even for streams that been already closed or at least inactive. That's been a logical consequence of PA's broken-by-design async concept for plugins20:12
SkryPA uses speex or ffmpeg (and perhaps something else) for conversion, speex can be a real cpu hog depending of the settings/method20:13
ShadowJKfreemangordon; that's why I was interested in stats on reads from compressed, if a large proportion is read from compressed cache it might be worthwhile to keep it.20:14
*** andre__ has quit IRC21:00
*** nox- has joined #maemo-ssu21:04
*** dafox has joined #maemo-ssu21:14
*** dafox is now known as Guest6786021:14
*** Guest67860 is now known as dafox21:21
*** dafox__ has joined #maemo-ssu21:45
*** dafox has quit IRC21:45
*** arcean has quit IRC22:15
*** arcean has joined #maemo-ssu22:18
*** Timo has joined #maemo-ssu22:24
*** Timo has quit IRC23:03
*** dafox__ has quit IRC23:32
*** arcean has quit IRC23:45
*** BCMM has quit IRC23:57

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!